INTRODUCTION

Type 1 diabetes mellitus (T1DM) is a complex and chronic disease that requires lifelong insulin injections, psychological support, and lifestyle changes. To optimize glycemic control, regular and frequent self- monitoring of blood glucose (SMBG) is required. Ziegler R, et al. found one additional SMBG per day to be associated with a decrease in HbA1c of 0.20%.1 Miller KM, el al. also found a higher number of SMBGs per day to be strongly associated with a lower HbA1c level.2 The numbers of SMBGs per day in the patients at our outpatient clinic3 was 2.7-3.3 times per day, which are less than the four to six times per day recommended by ISPAD clinical practice consensus guidelines.4 A possible reason for the inadequate number of SBMGs per day among our patients may be due to the high cost of the glucose test strips. For this reason, in 2015 our hospital organized “The Universal Coverage (UC) provided free glucose strips project for patients with T1DM”. However, fifteen months after initiation of this program, HbA1C was improved only in some patients in our clinic.

Another factor in addition to SMBG that contributes to good glycemic control is motivation. Motivational interviewing (MI), which was developed by Miller WR. and Rollnick S., is a proven approach for working through ambivalence and facilitating change of behavior.5 MI has been widely used in adults to improve control of addictive behaviors, such as reducing illicit drug use6 and promoting smoking cessation.7,8 During the last decade, MI has been used in pediatric practice to promote adherence to recommended treatment, including diabetes management with variable results in reducing HbA1C.9-12

MI is a brief, goal-directed, patient-centered counseling approach that was designed to help patients increase intrinsic motivation and strengthen commitment to change and improve via the exploration and resolution of ambivalence. Patients are encouraged to develop and recite their own self-motivational statement (SMS) by facilitators. The six stages of change in MI are described, as follows. The initial stage, which is labeled pre-contemplation, is when the person is not yet considering change. The next stage is the period of contemplation, during which the person evaluates the reasons for and against change. The third stage is when the person reaches a state of determination where plans for change are formulated. The person then takes action in the fourth stage to effectuate the identified change in behavior. If the change in behavior is successful, the person then moves into the fifth stage, which is a state of maintenance to sustain the change in behavior for the long term. The last of the six stages occurs if and when

the patient relapse, which is defined as a return to any of the previous behavior stages12

Thus we conducted a 6-month randomized controlled trial to evaluate the effectiveness of MI on glycemic control, as measured by HbA1C. The primary outcome was HbA1C at the 6-month follow-up. The secondary objective was to evaluate diabetes knowledge and self-care behavior. The secondary outcomes were the scores of the diabetes knowledge test and the self- care behavior questionnaire. We hypothesized that MI would improve glycemic control, diabetes knowledge, and self-care behavior in T1DM patients.

MATERIALS AND METHODS

Design and participants

Following randomization, participants received either diabetes self-management education (DSME) or DSME plus MI. Clinical staff and participants were both aware of the group assignment. Participants were recruited from the Outpatient Diabetes Clinic of the Division of Endocrinology and Metabolism, Department of Pediatrics, Siriraj Hospital during October 2016 to March 2017. Subjects were T1DM patients aged 10-18 years with HbA1C ≥8% that were receiving free glucose strips for at least 3 months. Patients who were receiving medications that effect glycemic control, such as steroids and switching of insulin regimen during this study, were excluded.

Randomization was generated by random permuted blocks with mixed block size. Group allocation results were sealed in sequentially numbered opaque envelopes. The person generated the allocation scheme had no additional role in the study. The protocol for this randomized controlled trial was approved by the Siriraj Institutional Review Board (SIRB) of the Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand (COA no. Si 538/2016). Assent and written informed consent was obtained from patients and their parents/guardians, respectively.

The frequency of SMBG was assessed by downloading glucometer data using accu-check 360º software. The information received were total numbers of SMBG in the past 3 months and average numbers of SMBG per day. Baseline characteristics and diabetes- related data including age, gender, insulin regimen, carbohydrate counting method, diabetes chronic complication, hypertension and dyslipidemia were collected. HbA1C levels were measured prior to entering the study and then at 3 and 6 months after entering the study.

Motivational interviewing (MI)

MI sessions were conducted by 3 interventionists, including 2 pediatric endocrinologists and a pediatric endocrinology fellow) and a diabetes education nurse. All interventionists were trained by experienced pediatric and adult psychiatrists from Department of Pediatrics and Department of Psychiatry, Faculty of Medicine Siriraj Hospital, Mahidol University. The initial training in MI includes hours of lecture, role play, case scenarios, and practicing with actual patients in individual and group sessions. Monthly discussion and supervision among interventionists and a psychiatrist was continued throughout the study.

MI in group session was performed at the beginning of the study, and at 3 months after entering the study. The length of the two MI sessions was 45-60 minutes each. MI by telephone call was performed individually at 1, 2, 4, and 5 months. Session dialogue included awareness building, making choices, alternatives, goal-setting, problem solving, and avoidance of confrontation. During MI sessions, interventionists encouraged patients to express self-motivational statements. The interventionists would respond to patients according to their stage of change. Interventionist responses included giving information and feedback for the pre-contemplation stage, discussion about pros and cons of undesired behavior for the contemplation stage, giving menu and promoting patient self-efficacy for the determination phase, encouraging compliance and adherence for the action stage, relapse prevention for the maintenance stage, and recovery process for the relapse stage. The MI manual was created by a pediatric psychologist. All MI sessions were documented, and all documentation was reviewed with a psychologist experienced in MI.

Diabetes self-management education (DSME) DSME in group session was performed at the

beginning of the study, and 3 months later in both the DSME and DSME plus MI groups. The session was designed as an interactive lecture and workshop, with a length of 60-90 minutes, and there were 8-10 patients in each class session. DSME consists of a diabetes knowledge component that was performed by physicians and a nurse, and a nutritional component that was performed by a nutritionist. Diabetes knowledge content included basic knowledge about diabetes, self-monitoring blood glucose, exercise with diabetes, hypo/hyperglycemia management, insulin action, sick-day management, and diabetes complications. Nutritional knowledge content included healthy food, carbohydrate-containing food, carbohydrate counting, food-exchange, and nutrition

facts. Food models were used for food exchange and nutrition fact practice. Patients were encouraged to participate in class by asking questions, giving examples, and using case scenarios. We also focused on individual problem-solving skills and insulin self-adjustment at home.

Diabetes knowledge test

Diabetes knowledge test was performed at the beginning and end of the study. We modified a multiple choice test using 30 questions from the diabetes knowledge test administered at the Siriraj Diabetes Camp.13 Questions covered 7 topics, including basic diabetes knowledge, nutritional management and carbohydrate counting, self-monitoring blood glucose, exercise with diabetes, hypo/hyperglycemia management, insulin treatment, and sick-day management.

Diabetes self-care behavior questionnaire Diabetes self-care behavior questionnaire was given

at the beginning and the end of the study. A 38-question standardized questionnaire that was developed by Tachanivate P.14 was used. The questionnaire covers 8 topics, including personal hygiene care, dietary control, medical taking, physical activity, self-monitoring blood glucose, problem solving, stress management, and reducing risk of diabetes complications. The score was reported as percentage of the mean, which was calculated using the following equation: % of mean = (actual sore/ maximum score) x 100. A higher score indicates better diabetes self-care behavior.

Statistical analysis

All data analyses were performed using SPSS Statistics (SPSS, Inc., Chicago, IL, USA). Patient characteristics were summarized using descriptive statistics. Categorical data were compared using chi- square test, and the results are presented as frequency or percentage. Normally distributed continuous data was compared using independent t-test, and the results were presented as mean ± standard deviation (SD). Non- normally distributed continuous data were compared using Mann-Whitney U-test, and the results were given as median and range (min, max). A p-value of less than 0.05 was considered statistically significant for all tests.

RESULTS

A flow diagram of the study protocol is shown in Fig 1. Of the 94 patients who received free glucose strips from the UC program, 39 were eligible for this study. Those patients were randomized into either the DSME

DSME (n=20)

-lost to follow up (n=1)

-switching insulin regimen (n=2)

At 3 months (n=17)

DSME with MI (n=19)

- drop out (n=1)

At 3 months (n=18)

At 6 months (n=17)

At 6 months (n=18)

group (n=20) or the DSME plus MI group (n=19). One patient in DSME group was lost to follow-up. One patient in the DSME plus MI group declined to participate after randomization, before the first visit. Two patients in the DSME group were excluded due to the fact that they switched insulin regimen during the study. The remaining 35 patients (17 DSME, and 18 DSME plus MI) completed the study.

Demographic characteristics

Patients in the DSME and DSME plus MI groups were well matched for age (14.18±2.02 vs. 14.06±2.88 years, respectively), age at diagnosis (8.25±2.86 vs. 8.53±3.83 years), duration of diabetes [5.25 (0.83, 13.33) vs. 5.08 (1, 14) years], and HbA1c [10.3% (8.4, 14) vs. 9.45% (8, 14.6)]. There were no significant differences between groups for age, age at diagnosis, duration of diabetes, or HbA1c, as shown in Table 1. Counting carbohydrate in grams was 30% and 22%; using basal

bolus regimen was 58% and 38% in the DSME and DSME plus MI groups respectively.

Primary outcome: HbA1C

Baseline HbA1C in the DSME and DSME plus MI groups was 10.3 (8.4, 14) and 9.45 (8, 14.6), respectively (p=0.204). At the end of the study, HbA1C in the DSME and DSME plus MI groups was 9.8 (7.4, 16.8) and 9.35 (7.8, 13.2), respectively (p=0.234). No significant difference was observed for HbA1C in each group compared between pre-intervention and post-intervention (Table 2).

Diabetes knowledge score

Diabetes knowledge score compared between baseline and 6 months increased significantly in both the DSME and DSME plus MI groups [19 (7, 24) to 21 (6, 25); p=0.012, and 18.5 (13, 24) to 21 (15, 28); p=0.001 respectively] (Table 3).

Self-care behavioral score

The self-care behavioral score was significantly different at 6 months compared between the DSME and DSME plus MI groups for the dietary control domain [48.97 (32.65, 73.47) vs. 60.20 (30.61, 77.55); p=0.024], and the medicine taking domain [52.38 (28.57, 78.57) vs. 67.86 (47.62, 80.95); p=0.016] (Table 4). There was no significant difference between groups at 6 months for the personal hygiene care, physical activity, self-monitoring,

problem solving, stress management, or reducing risk of diabetes complications domains.

Stage of MI

In DSME plus MI group, at the beginning of the study, there was 1 patient in pre-contemplation, 8 in contemplation, 9 in determination, and 0 in the action, maintenance, and relapse stages. At the end of the study, there were 12 patients in the action stage. (Table 5)

SMBG

Baseline SMBG frequency in the DSME and DSME plus MI groups was 2 (0.1, 4) and 3.2 (0.07, 4.9) times/day, respectively (p=0.013). At the end of the study, SMBG frequency in the DSME group and the DSME plus MI group was 2 (0, 4) and 3 (0, 4.7) times/day, respectively (p=0.053). SMBG frequency data was downloaded from the glucometer at baseline, 3- and 6-month time points.

DISCUSSION

We found no significant different in HbA1C between the DSME and DSME plus MI groups at the end of the study, as well as between pre- and post- intervention. Diabetes knowledge score in both groups was significantly increased at the end of study. Self-care behavioral score showed significant improvement in 2 domains (dietary control and medicine taking) in the DSME plus MI group. Transition to the action stage increased from 0 to 12 patients, and the transition occurred at approximately 4 months.

T1DM is a complex and chronic illness that requires consistent adherence to treatment, psychological support, and changes in lifestyle. Optimal glycemic control is not easy to achieve, requires commitment to change, and depends on multiple factors. Accurate carbohydrate counting is crucial for precise insulin calculation. The DAFNE Study Group reported significant improvement in HbA1C at 6 months (p<0.0001) after training patients how to match their insulin dose to their food choice.15 As demonstrated by Spiegel G, et al., T1DM patients overestimated and underestimated carbohydrate content, especially in mixed meals.16 Moreover, less than half of our patients were counting carbohydrates as grams, not portions (Table 1). Calculating insulin dose according to carbohydrate portion size may yield a loser insulin dose than calculating according to gram

weight. This may result in a suboptimal dose of insulin and poor glycemic control. Intensive patient education in carbohydrate counting and encouraging patients to count carbohydrates accurately may result in accurate insulin calculation and improving of glycemic control.

Non-intensive insulin regimen could be a barrier to achieving tight glycemic control. The American Diabetes Association (ADA) recommends that individuals with T1DM receive multiple daily insulin injections (three or more injections per day of prandial insulin, and one to two injections of basal insulin) or CSII.17 Hathout EH, et al. reported improvement in glycemic control with intensive therapy as compared with conventional insulin regimens.18 Only 58% and 38% of our patients in the DSME and DSME plus MI groups, respectively, used intensive insulin therapy, so tight glycemic control may be difficult to achieve. Likitmaskul S, et al. reported that Thai patients with T1DM had unsatisfactory glycemic control, with a mean HbA1C of 9.3±2.5%.19 Achieving good glycemic control in Thai patients may be challenging due to the fact that intensive diabetes treatment requires glucose test strips, and glucose strips are not available to all patients.

At the end of the intervention, the patients in the DSME plus MI group did better in the dietary control and medicine taking domains of self-care behavior than those in the DSME group, however, the HbA1C levels in the DSME plus MI group did not improved. This may be explained by the complexity of diabetes self-care, which requires multiple tasks of management. Their self-care behavior scores in other domains e.g. self-monitoring, problem solving, stress management, etc. were relatively low. No increase in frequency of SMBG and the fact that majority of patients in DSME plus MI group were treated with non-intensive insulin regimen might partly explain the lack of improvement in glycemic control. Moreover,

psychological issue might be another factor. The burden of having type 1 diabetes and the demands in managing daily diabetes-related tasks can lead to negative emotions or diabetes distress and depressive symptoms20 which can impact the glycemic control.

The duration of this study may have been too short to observe the effect of MI. In the present study, transition to the action stage of MI was observed at 4 months, so measurement of HbA1C at 6 months may be too early to observe the effect of action that recently took place. Channon S, et al. conducted a randomized controlled trial that showed significant reduction of HbA1C in the MI group compared to the control group at 12 months and 24 months, but not at 6 months.21 In our study, HbA1C was not significantly decreased in any comparison. On the other hand, MI may not affect glycemic control. Walter G. suggested that verbal indices of MI to change do not necessarily translate to actual change in response to treatment if the patient does not also have the ability to change, and that patient declarations should be regarded as reflecting the patient’s intent to change at that moment as opposed to being considered a predictor of real change in behavior.22

Diabetes knowledge score was significantly improved in both groups, which is similar to the finding reported by Santiprabhob, et al. at 6 months post-DSME at diabetes camp.23 Despite improving of diabetes knowledge score but the HbA1C levels did not improved may be due to the patients know the theory but did not apply the knowledge gained to daily life problem solving. From International Society for Pediatric and Adolescent Diabetes (ISPAD) recommendation and guidelines, educational interventions in children and adolescents with diabetes have a beneficial effect on both glycemic control and psychosocial outcomes.24 However, it is important to evaluate patients’ ability to apply their knowledge to their daily self-care.

The limitations of this study are short duration of intervention, infrequency of motivation intervention sessions and small sample size. Increasing the duration, intensity, and frequency of MI sessions, as well as focusing on individual ambivalence, may have positive impact on MI stage progression and actual change. It should also be considered that our small sample size may have given our study insufficient statistical power to identify all significant differences in HbA1C.

CONCLUSION

In conclusion, this study demonstrated that applying MI to diabetes care does not lead to improvement in glycemic control. However, diabetes knowledge was

improved in both groups, and self-care behavior score was improved in some topics. The process and methods for instilling and integrating diabetes knowledge, daily diabetes management, and self-care behavior, as well increasing the patient’s intrinsic motivation to change and improve, requires further study. Further study should also include a larger sample size, motivation that is focused on individualized specific issues, and a longer follow-up period.

ACKNOWLEDGEMENTS

The authors wish to thank the T1DM patients and their families that participated in this study, Julaporn Pooliam and Kanokwan Sommai for statistical analysis, and Hathaichanok Tirapongporn and Sriwan Thongpang for nutritional teaching. This study was supported by a Siriraj Routine to Research Fund, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand (grant no. R015935055).

Conflict of interest: The authors hereby declare no personal or professional conflicts of interest relating to any aspect of this study.

REFERENCES

1.Ziegler R, Heidtmann B, Hilgard D, Hofer S, Rosenbauer J, Holl R. Frequency of SMBG correlates with HbA1c and acute complications in children and adolescents with type 1 diabetes. Pediatr Diabetes 2011;12:11-7.

2.Miller KM, Beck RW, Bergenstal RM, Goland RS, Haller MJ, McGill JB, et al. Evidence of a strong association between frequency of self-monitoring of blood glucose and hemoglobin A1c levels in T1D exchange clinic registry participants. Diabetes Care 2013;36:2009-14.

3.Patjamontri S, Khemaprasit K, Santiprabhob J, Nakavachara P, Lertbannaphong O, Kiattisakthavee P, et al. The effect of early diabetes self-management education on glycemic control in children with type 1 diabetes. Southeast Asian J Trop Med Public Health 2018;49:304-13.

4.Rewers MJ, Pillay K, de Beaufort C, Craig ME, Hanas R, Acerini CL, et al. ISPAD Clinical Practice Consensus Guidelines 2014. Assessment and monitoring of glycemic control in children and adolescents with diabetes. Pediatr Diabetes 2014;15 Suppl 20:102-14.

5.Miller WR, Rollnick S. Meeting in the middle: motivational interviewing and self-determination theory. Int J Behav Nutr Phys Act 2012;9:25.

6. Li L, Zhu S, Tse N, Tse S, Wong P. Effectiveness of motivational interviewing to reduce illicit drug use in adolescents: a systematic review and meta-analysis. Addiction 2016;111:795-805.

7.Lai DT, Cahill K, Qin Y, Tang JL. Motivational interviewing for smoking cessation. Cochrane Database Syst Rev 2010:CD006936.

8.Lindson-Hawley N, Thompson TP, Begh R. Motivational interviewing for smoking cessation. Cochrane Database Syst Rev 2015:CD006936.

9.Erickson SJ, Gerstle M, Feldstein SW. Brief interventions and

motivational interviewing with children, adolescents, and their parents in pediatric health care settings: a review. Arch Pediatr Adolesc Med 2005;159:1173-80.

10.Jones A, Gladstone BP, Lubeck M, Lindekilde N, Upton D, Vach

W.Motivational interventions in the management of HbA1c levels: a systematic review and meta-analysis. Prim Care Diabetes 2014;8:91-100.

11.Sindelar HA, Abrantes AM, Hart C, Lewander W, Spirito A. Motivational interviewing in pediatric practice. Curr Probl Pediatr Adolesc Health Care 2004;34:322-39.

12.Suarez M, Mullins S. Motivational interviewing and pediatric health behavior interventions. J Dev Behav Pediatr 2008;29:417- 28.

13.Santiprabhob J, Kiattisakthavee P, Likitmaskul S, Chaichanwattanakul K, Wekawanich J, Dumrongphol H, et al. Glycemic control, quality of life and self-care behavior among adolescents with type 1 diabetes who attended a diabetes camp. Southeast Asian J Trop Med Public Health 2012;43:172-84.

14.Tachanivate P. Factors influencing quality of life in adolescents with type 1 diabetes [Thesis]. Bangkok, Thailand: Mahidol University; 2007.

15.DAFNE Study Group. Training in flexible, intensive insulin management to enable dietary freedom in people with type 1 diabetes: dose adjustment for normal eating (DAFNE) randomised controlled trial. BMJ 2002;325:1-6.

16.Spiegel G, Bortsov A, Bishop FK, Owen D, Klingensmith GJ, Mayer-Davis EJ, et al. Randomized nutrition education intervention to improve carbohydrate counting in adolescents with type 1 diabetes study: is more intensive education needed?

JAcad Nutr Diet 2012;112:1736-46.

17.Chiang JL, Kirkman MS, Laffel LM, Peters AL. Type 1 diabetes

Original Article SMJ

through the life span: a position statement of the American Diabetes Association. Diabetes Care 2014;37:2034-54.

18.Hathout EH, Fujishige L, Geach J, Ischandar M, Maruo S, Mace JW. Effect of therapy with insulin glargine (lantus) on glycemic control in toddlers, children, and adolescents with diabetes. Diabetes Technol Ther 2003;5:801-6.

19.Likitmaskul S, Wacharasindhu S, Rawdaree P, Ngarmukos C, Deerochanawong C, Suwanwalaikorn S, et al. Thailand diabetes registry project: type of diabetes, glycemic control and prevalence of microvascular complications in children and adolescents with diabetes. J Med Assoc Thai 2006;89 Suppl 1:S10-6.

20.Dejkhamron P, Santiprabhob J, Likitmaskul S, Deerochanawong

John RL, Rogers C, et al. A multicenter randomized controlled trial of motivational interviewing in teenagers with diabetes. Diabetes Care 2007;30:1390-5.

22.Waller G. The myths of motivation: time for a fresh look at some received wisdom in the eating disorders? Int J Eat Disord 2012;45:1-16.

23.Santiprabhob J, Likitmaskul S, Kiattisakthavee P, Weerakulwattana P, Chaichanwattanakul K, Nakavachara P, et al. Glycemic control and the psychosocial benefits gained by patients with type 1 diabetes mellitus attending the diabetes camp. Patient Educ Couns 2008;73:60-6.

24.Lange K, Swift P, Pankowska E, Danne T. ISPAD Clinical Practice Consensus Guidelines 2014. Diabetes education in children and adolescents. Pediatr Diabetes 2014;15 Suppl 20:77-85.